US20140320212A1 - Radio frequency power amplifiers - Google Patents
Radio frequency power amplifiers Download PDFInfo
- Publication number
- US20140320212A1 US20140320212A1 US14/356,677 US201214356677A US2014320212A1 US 20140320212 A1 US20140320212 A1 US 20140320212A1 US 201214356677 A US201214356677 A US 201214356677A US 2014320212 A1 US2014320212 A1 US 2014320212A1
- Authority
- US
- United States
- Prior art keywords
- transistor
- input
- output
- active device
- power amplifier
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000003990 capacitor Substances 0.000 claims description 35
- 230000003071 parasitic effect Effects 0.000 claims description 28
- 238000000034 method Methods 0.000 claims description 23
- 239000004065 semiconductor Substances 0.000 claims description 11
- 230000005669 field effect Effects 0.000 claims description 10
- 229910044991 metal oxide Inorganic materials 0.000 claims description 5
- 150000004706 metal oxides Chemical class 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 3
- 229910052751 metal Inorganic materials 0.000 claims description 3
- 239000002184 metal Substances 0.000 claims description 3
- 230000008569 process Effects 0.000 description 14
- 230000006835 compression Effects 0.000 description 10
- 238000007906 compression Methods 0.000 description 10
- 229910000577 Silicon-germanium Inorganic materials 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000013459 approach Methods 0.000 description 4
- LEVVHYCKPQWKOP-UHFFFAOYSA-N [Si].[Ge] Chemical compound [Si].[Ge] LEVVHYCKPQWKOP-UHFFFAOYSA-N 0.000 description 3
- 230000003321 amplification Effects 0.000 description 3
- 238000003199 nucleic acid amplification method Methods 0.000 description 3
- 239000000758 substrate Substances 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- JBRZTFJDHDCESZ-UHFFFAOYSA-N AsGa Chemical compound [As]#[Ga] JBRZTFJDHDCESZ-UHFFFAOYSA-N 0.000 description 2
- 230000000295 complement effect Effects 0.000 description 2
- 230000002939 deleterious effect Effects 0.000 description 2
- 238000011982 device technology Methods 0.000 description 2
- 229920006395 saturated elastomer Polymers 0.000 description 2
- 229910001218 Gallium arsenide Inorganic materials 0.000 description 1
- JMASRVWKEDWRBT-UHFFFAOYSA-N Gallium nitride Chemical compound [Ga]#N JMASRVWKEDWRBT-UHFFFAOYSA-N 0.000 description 1
- GPXJNWSHGFTCBW-UHFFFAOYSA-N Indium phosphide Chemical compound [In]#P GPXJNWSHGFTCBW-UHFFFAOYSA-N 0.000 description 1
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000008713 feedback mechanism Effects 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
- H03F3/189—High-frequency amplifiers, e.g. radio frequency amplifiers
- H03F3/19—High-frequency amplifiers, e.g. radio frequency amplifiers with semiconductor devices only
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F1/00—Details of amplifiers with only discharge tubes, only semiconductor devices or only unspecified devices as amplifying elements
- H03F1/34—Negative-feedback-circuit arrangements with or without positive feedback
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
- H03F3/20—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers
- H03F3/21—Power amplifiers, e.g. Class B amplifiers, Class C amplifiers with semiconductor devices only
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F2200/00—Indexing scheme relating to amplifiers
- H03F2200/138—Indexing scheme relating to amplifiers the feedback circuit comprising a parallel resonance circuit
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F2200/00—Indexing scheme relating to amplifiers
- H03F2200/451—Indexing scheme relating to amplifiers the amplifier being a radio frequency amplifier
Definitions
- the subject matter described herein relates to radio frequency power amplifiers.
- radio frequency applications including automotive radar systems, terrestrial wireless communications, and satellite communications require radio frequency power amplifiers.
- Some applications requiring power amplifiers operate at millimeter-wave frequencies. At high frequencies including millimeter-wave frequencies, designing and producing power amplifiers with high output power and high efficiency is challenging.
- Some other challenges include input impedance matching to a source impedance, and output impedance matching to a load impedance. These challenges are due in part to the deleterious effects of unavoidable parasitic elements of the active devices used to create amplification. For example, the parasitic base-collector capacitance and the parasitic base-emitter capacitance can degrade the input and output matching, amplifier efficiency, and output power.
- Methods and apparatus including computer program products, are provided for a circuit and a circuit model of a radio frequency power amplifier providing high power output, high efficiency, and good input and output matching at radio frequencies including millimeter-wave frequencies.
- an apparatus may include a first transmission line carrying a signal to an input port of a power amplifier, the power amplifier producing an output signal at an output port of the power amplifier.
- the power amplifier may include one or more power amplifier stages.
- a power amplifier stage may include an active device, and a feedback network comprising one or more reactive elements configured to resonate at a predetermined frequency.
- the feedback network may also be configured to provide a impedance matching at an input to the amplifier stage, and to provide impedance matching at an output of the amplifier stage, the input impedance matching and the output impedance matching configured by at least biasing the active device to produce a transconductance at least one of equal to or greater than a critical transconductance.
- the one or more active devices may include a bipolar junction transistor, a field effect transistor, a metal oxide semiconductor field effect transistor, a metal semiconductor field effect transistor, and a high electron mobility transistor.
- a method may be provided. The method may include determining a bias point for an active device in a radio frequency power amplifier, wherein the bias point is selected to cause a transconductance of the active device to be above a critical transconductance value; determining a parasitic capacitance between an input of the active device and an output of the active device; selecting one or more reactive components in a feedback circuit of the radio frequency power amplifier to produce a parallel tank circuit, the one or more reactive components including the parasitic capacitance; and determining one or more values for the one or more reactive components in the parallel tank circuit to cause an input impedance match and an output impedance match.
- the feedback network may include an inductor connected between an input port and an output port, an input capacitor connected between the input port and the base of the transistor, a parasitic capacitance of the transistor between the base of the transistor and the collector of the transistor, and an output capacitor connected between the output port and the collector of the transistor.
- the feedback network may include a feedback capacitor in parallel with the parasitic base-collector capacitance of the transistor, and/or a parasitic capacitance of the transistor between the base of the transistor and the emitter of the transistor.
- FIG. 1 is a diagram showing a source, a load, and an example of a power amplifier, in accordance with some implementations
- FIG. 2A depicts a diagram showing an example of a single stage radio frequency (RF) power amplifier incorporating an active device, in accordance with some implementations;
- RF radio frequency
- FIG. 2B depicts a diagram showing an example of a single stage RF power amplifier incorporating a bipolar junction transistor as the active device, in accordance with some implementations;
- FIG. 3A shows an alternating circuit (AC) circuit diagram of a single stage RF power amplifier, in accordance with some implementations
- FIG. 3B shows a small-signal AC circuit model including a feedback network, in accordance with some implementations
- FIG. 4 shows a process for producing an RF power amplifier stage, in accordance with some implementations
- FIG. 5A shows the two-port S-parameters of a circuit similar to FIG. 2B ;
- FIG. 5B shows the effect of the voltage divider created by capacitances, such as input capacitance Ci and base-emitter capacitance C ⁇ , on the collector current as a function of input power, in accordance with some implementations;
- FIG. 6 is a circuit diagram showing a three-stage power amplifier, in accordance with some implementations.
- FIG. 7 is a photograph of a chip containing a three-stage power amplifier, in accordance with some implementations.
- FIG. 1 depicts an example of a radio frequency power amplifier 150 including a radio frequency (RF) source 110 having a source impedance Z s 120 , a feedback network 190 , an active device 195 , and a load 180 having a load impedance Z L 180 , in accordance with some example implementations.
- the RF source 110 may have an input power P i 130 available to supply to RF power amplifier 150 at input port 135 .
- the RF source 110 may provide a radio frequency carrier that is modulated to carry information.
- the RF source 110 may generate a RF carrier having a frequency between 5 GHz and 150 GHz including the millimeter-wave bands, such as Q-band (30-50 GHz), V-band (50-75 GHz), and W-band (75-110 GHz), although other frequency ranges may be used as well.
- the radio frequency amplifier 150 may also include an input impedance Z in 140 (e.g., in ohms), an output power P o 170 delivered at output port 165 to load impedance Z L 180 , and an output impedance Z out 160 (e.g., in ohms).
- RF power amplifier 150 amplifies input power P i 130 and delivers output power P o to load impedance Z L 180 .
- RF power amplifier circuit 100 may be configured, such that a source impedance, an input impedance, an output impedance, and a load impedance are each determined so as to substantially match an impedance value (referred to herein as a characteristic impedance).
- a characteristic impedance For example, source impedance Z s 120 and load impedance Z L 180 may each be selected to have a value equal to about a characteristic impedance, Z o .
- RF power amplifier 150 may be configured so that active device 195 and feedback network 190 are selected to provide input impedance Z in 140 equal to about the characteristic impedance and output impedance Z out 160 equal to about the characteristic impedance.
- the characteristic impedance may be equal to about 50 ohms, although other impedance values may be used as well.
- the feedback network 190 combined with the active device 195 may be configured to nullify the deleterious effects of the parasitic elements of active device 195 in order to provide higher output power, greater amplifier efficiency, and/or provide good input and output matching.
- the combination of the input impedance Z in 140 and the source impedance Z s 120 may determine a reflection coefficient (related to as the Scattering parameter S 11 , when the input port 135 is port 1 ).
- the reflection coefficient is about equal to zero or nearly zero (e.g., less than or equal to about 0.1 or equivalently ⁇ 10 dB)
- the impedances such as source impedance Z s 120 and RF power amplifier input impedance Z in , may be considered well matched.
- the parasitic elements of one or more active devices may be combined with external components, such as for example inductors and capacitors selected to form a tank circuit resonant at a predetermined frequency.
- the tank circuit may include parasitic components of the active device 195 , such as a parasitic base-collector capacitance, C ⁇ , if the active device is a bipolar junction transistor.
- the RF power amplifier 150 may, in some example implementations, achieve a higher output power before going into gain compression, when compared to an active device without the parallel tank circuit.
- Transconductance values above the critical value may also provide higher output power before going into compression.
- selecting a transconductance value below the critical value may result in poorer input and output matching and degraded linearity and output compression at a lower output power compared to the matching and linearity when the transconductance is above the critical value.
- the feedback network 190 and the active device 195 may also cause an impedance match at the input port 135 between the input impedance Z in 140 and source impedance Z s 120 .
- the feedback network 190 and the active device 195 having a transconductance at, or above, the threshold value may cause an impedance match at the output port 165 between the output impedance Z out 160 and load impedance Z L 180 .
- the RF amplifier 150 may include one or more active devices and one or more passive devices.
- active devices include a bipolar junction transistor (BJT), a heterojunction bipolar transistor (HBT), a field effect transistor (FET), a metal oxide semiconductor field effect transistor (MOSFET), a metal semiconductor field effect transistor (MESFET), and a high electron mobility transistor (HEMT), although other active devices may be included in RF amplifier 150 as well.
- the active device 195 is chosen to be a bipolar junction transistor produced in a 120 nm silicon-germanium (SiGe) BiCMOS process, although other processes may be used as well.
- Examples of passive devices include an inductor, a capacitor, and a transmission line, although other types of passive devices may be included in RF power amplifier 150 as well.
- the devices used in the power amplifier 150 may be implemented in a semiconductor processes, such as for example silicon (Si), silicon germanium (SiGe), gallium arsenide (GaAs), indium phosphide (INP), gallium nitride (GaN) as well as others.
- the semiconductor process may allow for multiple active device types, such as in a bipolar/complementary metal oxide semiconductor (BiCMOS) process that allows for biploar transistors and complementary metal oxide semiconductor (CMOS) devices to be produced using the same process.
- BiCMOS bipolar/complementary metal oxide semiconductor
- RF power amplifier 150 may include a single amplification stage, such as amplifier stage 152 or multiple amplifier stages. Multiple amplifier stages may be cascaded with the input port 135 connected to the input to the first stage and the output of the first stage being connected to the input of the next stage, and so on, up to the last stage where the output of the last stage is connected to output port 165 . Multiple amplifier stages may also be interconnected in parallel where the outputs of multiple stages are combined together at output port 165 . A combination of cascaded amplifier stages and parallel amplifier stages may also be used.
- FIG. 2A depicts an example of a single RF power amplifier stage 200 A.
- the RF power amplifier single stage 200 A may include an input port 260 for receiving a signal for amplification, a feedback network 210 , an active device 220 A, and an output port 270 for supplying an amplified output signal.
- the active device 220 A may be an active device configured in accordance with a semiconductor process as noted above (e.g., a bipolar junction transistor implemented in a BiCMOS process, and the like).
- the feedback network 210 may include an input capacitor C i 230 , an inductor L 240 , an output capacitor C o 250 , and feedback capacitor C ⁇ est 255 .
- input capacitor C i 230 may connect input port 260 to the gate/base/input of active device 220 A
- inductor L 240 may connect input port 260 to output port 270
- output capacitor C o 250 may connect output port 270 to the drain/collector/output of active device 220 A
- feedback capacitor C ⁇ ext 255 may connect the input port 260 to the output port 270 .
- the source/emitter of active device 200 A may be connected to a ground (e.g., an alternating current (AC) ground).
- a ground e.g., an alternating current (AC) ground
- FIG. 2B depicts an example of a single RF power amplifier stage 200 B implemented using a bipolar junction transistor 220 B.
- input capacitor C i 230 may connect input port 260 to the base of bipolar junction transistor 220 B
- inductor L 240 may connect input port 260 to output port 270
- output capacitor C o 250 may connect output port 270 to the collector of bipolar junction transistor 220 B
- feedback capacitor C ⁇ ext 255 may connect the input port 260 to the output port 270 .
- the parasitic capacitance of bipolar junction transistor 220 B may also connect the input port 260 to the output port 270 .
- the emitter of bipolar junction transistor 220 B may be connected to AC ground.
- a parallel tank circuit may be formed from inductor L 240 in parallel with the series combination of input capacitance C i 230 , output capacitance C o 250 , and the parallel combination of parasitic base-collector capacitance C ⁇ of bipolar junction transistor 220 B in parallel with feedback capacitance C ⁇ ext ext 255 .
- feedback capacitance C ⁇ ext 255 may have a zero or nearly zero capacitance value.
- the parasitic base-collector capacitance C ⁇ of bipolar junction transistor 220 B may be an equivalent value determined based on the configuration of bipolar junction transistor 220 B and the process selected to fabricate bipolar junction transistor 220 B.
- the value of parasitic base-collector capacitance C ⁇ may be determined based on the device physics, or measured.
- FIG. 3A depicts an example of a single stage RF power amplifier 300 , which is similar in some respects to the power amplifier 200 B in FIG. 2B .
- the amplifier 300 may include an input source 110 with a source impedance Z s 120 about equal to a characteristic impedance Z o and a load impedance 180 about equal to a characteristic impedance Z o .
- FIG. 3 A depicts an AC equivalent circuit without a bias network, although it may be presumed that the bipolar transistor 220 B is biased to a transconductance above a critical transconductance.
- FIG. 3B depicts an example of a small-signal circuit model 305 representation of the single stage RF power amplifier 300 at FIG. 3A .
- the small-signal circuit model 305 may include the active device model parameters, such as parasitic base-collector capacitance C ⁇ 315 , transconductance g m 330 , substrate capacitance C s 340 , and base-emitter capacitance C ⁇ 320 .
- the parallel resonant tank circuit 310 may include a feedback network, such as feedback network 210 and the parasitic base-collector capacitance (or gate-drain capacitance) of the active device, such as base-collector capacitance C ⁇ 315 .
- g m When the active device transconductance, g m , is selected to be greater than the critical transconductance (e.g., g m ⁇ g m,crit ), the following device behavior may be observed:
- C eff may be determined in accordance with the following:
- C ⁇ . and C ⁇ ext are referred to as just C ⁇ .
- the voltage gain, A v represents the voltage gain of the power amplifier, wherein A v may be determined in accordance with the following:
- the transconductance 330 of the active device 220 B and the parallel tank circuit 310 may cause the input impedance Z in 140 to be about equal to the characteristic impedance Z o and may cause the output impedance Z out 160 to be about equal to characteristic impedance Z o .
- the two-port S-parameters of the small-signal circuit model shown in FIG. 3B are as follows:
- the input and output return loss, S 11 and S 22 are expressed in Equation 3A.
- the tank circuit including feedback network 210 and active device 220 A/B may be configured to provide both input and output matching.
- ⁇ notch ⁇ square root over ((1)/( C eff L ) ⁇ (1)/( C i Z o ) 2 ) ⁇ square root over ((1)/( C eff L ) ⁇ (1)/( C i Z o ) 2 ) ⁇ square root over ((1)/( C eff L ) ⁇ (1)/( C i Z o ) 2 ) ⁇ square root over ((1)/( C eff L ) ⁇ (1)/( C i Z o ) 2 ) ⁇ Equation 3.
- ⁇ notch approaches ⁇ o .
- the input impedance is high compared to characteristic impedance Z o , since ⁇ dominates both the numerator and denominator in Equation 2A.
- the parallel tank circuit 310 and transconductance 330 simplify to a parallel LC circuit which is an open circuit at resonant frequency ⁇ o .
- the real part of the input (and output) impedance decreases and the input and output impedances converge to the following:
- the input and output matching may, in some example implementations, improve as the input capacitance C i 230 becomes larger.
- the required transconductance g m for a return loss better than 10 dB may be defined as the critical transconductance as follows:
- Increasing both the transconductance g m and the ratio of input capacitance to the base-collector capacitance may, in some example implementations, improve the return loss.
- C i /C ⁇ the ratio of input capacitance to the base-collector capacitance
- increasing the transconductance g m above the critical transconductance value g m,crit may not substantially improve (e.g., increase) the return loss.
- the frequency where S 11 and S 22 are at a minimum (which is also referred to as a notch frequency, ⁇ notch ) and the frequency where S 21 peaks, ⁇ peak , are related to the resonant frequency ⁇ o as follows:
- Equations 2B and 2C may be simplified as follows:
- the quality factor, Q may be expressed as follows:
- the quality factor Q of the of the parallel tank circuit in Equation 8 may be inversely proportional to the value of the inductor L 240 , a smaller inductor value may make the parallel tank more narrowband, while a larger value may make the parallel tank more broadband.
- a smaller value of the effective capacitance C eff may make the parallel tank more wideband, while a larger capacitor value may make the parallel tank more narrowband.
- the return losses, S 11 and S 22 may be broadened using balanced architectures or staggering the return loss across multiple power amplifier stages.
- the above disclosure is based on the base-emitter capacitance C ⁇ and the substrate capacitance C s being zero or about zero.
- the substrate capacitance C s may be relatively small and thus has minor effects.
- the base-emitter capacitance C ⁇ may be an order of magnitude larger than the input capacitance C i due to the value of the transconductance g m .
- the base-emitter capacitance C ⁇ reduces the gain, S 21 . However, this reduction in gain may have the beneficial effect of keeping the active device out of compression at higher output voltage swings.
- FIG. 4 depicts a process 400 for producing an RF power amplifier stage, in accordance with some example implementations.
- a processor may determine, for the RF amplifier, one or more of the following amplifier parameters: a device technology (e.g., silicon-germanium, gallium-arsenide, etc), an active device structure (e.g., bipolar junction transistor, field-effect transistor, etc.), a quantity of power amplifier stages, an operating frequency, a gain, an output power, and any other parameters which may be determined or optimized by a processor. Some of these parameters may be dictated based on, for example, the intended application, such as a satellite communications application for a particular satellite. For example, an RF power amplifier for millimeter-wave satellite communications may require 200 milliwatts (23 dBm) of output power at 38 Gigahertz.
- the parameters which may be determined at 410 may thus include the device technology, active device structure, the quantity of stages, and the gain.
- a bias point for the active device such as active deice 195
- the active device's transconductance (which is determined by the bias point) is about greater than a critical transconductance value, as described above with respect to FIGS. 3A and 3B and Equation 5.
- a processor may select a bias point representative of a transconductance value greater than the critical value given in Equation 5.
- the processor may select the bias point by selecting the bias voltage that corresponds to the selected transconductance.
- selecting a transconductance value below the critical value may result in poorer input and output matching and/or degraded linearity compared to the matching and linearity realized when the transconductance is above the critical value.
- the bias point may be implemented by setting a direct current (DC) base/source/input voltage.
- the values of the components of a circuit model of the active device may be determined.
- the values of the base-emitter capacitance C ⁇ 320 , and the base-collector capacitance C ⁇ 315 may depend on the bias point of bipolar junction transistor 220 B.
- the values of the base-emitter capacitance C ⁇ 320 and C ⁇ 315 may be determined after the bias point is determined.
- the values of the components of the feedback network such as feedback network 210 with components including input capacitance C i 230 , output capacitance C o 250 , inductance L 240 , and feedback capacitance C ⁇ ext 255 may be chosen.
- the values of these components may be chosen to provide, when the active device transconductance 330 is substantially above the critical transconductance in accordance with Equation 5, a good input match at input port 260 , a good output match at output port 270 , and a parallel tank 310 with a predetermined resonant frequency, ⁇ o .
- a processor may select input capacitance C i 230 , output capacitance C o 250 , inductance L 240 , feedback capacitance C ⁇ ext 255 , and a bias point to provide a transconductance above the critical transconductance in accordance with Equation 5, and good input and output matching in accordance with Equation 2A.
- the feedback network includes the parallel tank including inductor L 240 in parallel with the series combination of input capacitance C i 230 , base-collector capacitance C ⁇ 315 from the bipolar junction transistor small-signal circuit model, output capacitance C o 250 , and feedback capacitance C ⁇ ext 255 .
- the predetermined resonant frequency of the parallel tank circuit is the center of the operating frequency of the RF power amplifier.
- process 400 may be repeated for each stage of a multistage RF power amplifier and/or repeated for an individual stage. In any case, process 400 may be repeated in order to optimize the determination of the active bias point and the feedback network.
- the large-signal behavior of RF power amplifier stage 300 in FIG. 3A may depend on the biasing of bipolar junction transistor 220 B.
- the circuit in FIG. 3B may exhibit one or more of the following attributes at the resonant frequency of the parallel tank circuit 310 :
- the collector efficiency ⁇ may be represented as a ratio of the output power P out 170 to the direct current power supplied to the power amplifier P dc or in terms of the peak output current i o , characteristic impedance Z o , direct current collector bias voltage V cc , and collector current I c according to the following equation:
- the maximum collector efficiency may be expressed in terms of the voltage gain A v as follows:
- the efficiency, ⁇ , in Equation 9 may be influenced by the RF power amplifier class (a class A amplifier, a class B amplifier, a class AB amplifier, and the like), the RF power amplifier 150 may achieve, in some example implementations, higher efficiency at high input power levels because in part the load line matching does not require a high-Q impedance transformation.
- tank circuit 310 may provide both input and output matching.
- the base-emitter capacitance, C ⁇ 320 may substantially linearize the power amplifier input-output transfer function, which may provide for a higher input power before the output goes into compression. This linearization may be due to a voltage divider formed by input capacitor C i 230 and base-emitter capacitance C ⁇ 320 , wherein input voltage V i is the input voltage at input port 260 and base-emitter voltage V ⁇ is the voltage between the base and emitter of bipolar junction transistor 220 B.
- the voltage divider causes a portion of the input voltage V i to be across the input capacitor C i and the remaining portion of the input voltage V i to be across the base-emitter (equal to V ⁇ ) of bipolar junction transistor 220 B.
- the base-emitter voltage V ⁇ may be represented in terms of a ratio involving the input capacitance C i 230 and the base-emitter capacitance C ⁇ 320 in Equation 11. With the base-collector capacitance C ⁇ 315 and feedback capacitance C ⁇ ext 255 in FIG. 3B being small compared to C i and C ⁇ , they may be neglected in the representation of the voltage divider.
- FIG. 5A shows a two-port S-parameters of a circuit similar to FIG. 2B . Shown are the input match S 11 , the output match S 22 , the gain S 21 and the reverse gain S 12 . The minimum of S 11 and S 22 at the notch frequency, ⁇ notch 502 , related to Equations 3 and 6 is also shown. The maximum of S 21 at the peak frequency , ⁇ peak 501 , related to Equation 6 is also shown.
- FIG. 5B shows the effect of the voltage divider.
- the voltage divider causes the RF power amplifier disclosed here to delay compression of the output signal at output port 270 to a higher power input signal at input port 260 compared to traditional design techniques.
- the plot at 510 depicts the bipolar junction transistor collector current as a function of input power, in accordance with some example implementations.
- the voltage V ⁇ 510 A at the base of the active device, such as bipolar junction transistor 220 B, may be the result of the input voltage V i divided across input capacitance C i and base-emitter capacitance C ⁇ wherein,
- V ⁇ c i c i + c ⁇ ⁇ V i . Equation ⁇ ⁇ 11
- the effect of the voltage divider may be to linearize the transfer function of the output current as a function of input power. This may extend the range of input power before the output of the power amplifier goes into compression.
- the effect of the voltage divider may be to increase the input power corresponding to the 1 dB compression point of the output 270 .
- the voltage divider may, in some example implementations, improve the 1 dB compression point by a factor involving the ratio of the base-emitter capacitance to the input capacitance expressed as follows:
- the plot at 520 depicts the input power versus the collector current without input capacitor C i and thus no voltage divider. Without input capacitor C i , the voltage across the base-emitter V ⁇ 520 A of the transistor is equal to the input voltage V i . The result of not having the voltage divider is to produce more non-linearity in the input power versus collector current transfer function thus reducing the input power corresponding to the 1 dB compression point of the output.
- FIG. 6 is a circuit diagram showing a three-stage power amplifier 600 .
- Stage 1 at 610 receives an input signal from input port 135 through inductor L 1 611 .
- the stage 1 parallel tank circuit includes inductor L 2 612 , capacitor C 1 613 , capacitor C 2 615 and the base-collector capacitance (C ⁇ ) of bipolar junction transistor 616 .
- Bipolar junction transistor 616 is biased at 618 by a bias network such as bias network 601 .
- inductor L 1 150 pH (picohenries)
- inductor L 2 320 pH
- bipolar junction transistor 616 has a size of 36 microns.
- Stage 2 at 620 receives an input signal from stage 1 through inductor L 3 621 .
- the stage 2 parallel tank circuit includes inductor L 4 622 , capacitor C 3 623 , capacitor C 4 625 and the base-collector capacitance (C ⁇ ) of bipolar junction transistor 626 .
- Bipolar junction transistor 626 is biased at 628 by a bias network such as bias network 601 .
- inductor L 3 150 pH
- inductor L 4 240 pH
- bipolar junction transistor 626 has a size of 72 microns.
- Stage 3 at 620 receives an input signal from stage 2 through inductor L 5 631 .
- the stage 3 parallel tank circuit includes inductor L 6 632 , capacitor C 5 633 , capacitor C 6 635 and the base-collector capacitance (C ⁇ ) of bipolar junction transistor 636 .
- Bipolar junction transistor 636 is biased at 638 by a bias network such as bias network 601 .
- inductor L 5 150 pH
- inductor L 6 180 pH
- bipolar junction transistor 626 has a size of 144 microns.
- Output port 165 provides the output of RF power amplifier 600 from stage 3 through inductor L 7 641 which has a value of 150 pH.
- the component values in power amplifier 600 are representative of a particular implementation where the values a chosen in accordance with the foregoing disclosure. Any number of other implementations are also possible that satisfy the foregoing considerations.
- FIG. 7 shows a drawing 700 of a die containing a circuit similar to the circuit of FIG. 6 implemented using a 120 nm SiGe BiCMOS process.
- the circuit measures 1160 microns ⁇ 900 microns.
- the implementation at 700 has three RF power amplifier stages 610 , 620 , and 630 , which are similar in some respects to the three stages at FIG. 6 .
- the input port to the RF power amplifier in 700 is a coplanar waveguide ground-signal-ground structure shown in 135 A and 135 B.
- the physical inductors of FIG. 6 are depicted as inductors L 1 611 , L 2 612 , L 3 621 , L 4 622 , L 5 631 , L 6 632 and L 7 641 .
- the power amplifier output port is depicted at 165 A and 165 B.
- Capacitors C 1 613 , C 2 615 , C 3 623 , C 4 625 , C 5 633 , and C 6 635 are also present but not labeled as are bipolar junction transistors 616 , 626 , and 636 .
- the nominal collector current bias for all three stages is 4 mA at 2.4-V collector voltage.
- This RF power amplifier built in accordance with the foregoing may, in some example implementations, achieve a maximum power added efficiency of 20% with a saturated output power of 21.1 dBm; a saturated output power of 23 dBm with a collector voltage of 3 V; and/or a 3-dB bandwidth of 5-GHz from 36 to 41 GHz, although other implementations may realize other performance values as well.
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Amplifiers (AREA)
Abstract
Description
- This application claims the benefit of U.S. Provisional Patent Application Ser. No. 61/557,866, filed on Nov. 9, 2011, and entitled “FEEDBACK MECHANISM FOR HIGH POWER RF POWER AMPLIFIERS,” which is incorporated by reference herein in its entirety.
- Certain aspects of the subject matter described herein were developed with U.S. Government support under Grant No. W911NF-10-1-0089 awarded by ARMY-ARO. The U.S. Government has certain rights in the invention.
- The subject matter described herein relates to radio frequency power amplifiers.
- Many radio frequency applications including automotive radar systems, terrestrial wireless communications, and satellite communications require radio frequency power amplifiers. Some applications requiring power amplifiers operate at millimeter-wave frequencies. At high frequencies including millimeter-wave frequencies, designing and producing power amplifiers with high output power and high efficiency is challenging. Some other challenges include input impedance matching to a source impedance, and output impedance matching to a load impedance. These challenges are due in part to the deleterious effects of unavoidable parasitic elements of the active devices used to create amplification. For example, the parasitic base-collector capacitance and the parasitic base-emitter capacitance can degrade the input and output matching, amplifier efficiency, and output power.
- Methods and apparatus, including computer program products, are provided for a circuit and a circuit model of a radio frequency power amplifier providing high power output, high efficiency, and good input and output matching at radio frequencies including millimeter-wave frequencies.
- In some example embodiments, an apparatus is provided. The apparatus may include a first transmission line carrying a signal to an input port of a power amplifier, the power amplifier producing an output signal at an output port of the power amplifier. The power amplifier may include one or more power amplifier stages. A power amplifier stage may include an active device, and a feedback network comprising one or more reactive elements configured to resonate at a predetermined frequency. The feedback network may also be configured to provide a impedance matching at an input to the amplifier stage, and to provide impedance matching at an output of the amplifier stage, the input impedance matching and the output impedance matching configured by at least biasing the active device to produce a transconductance at least one of equal to or greater than a critical transconductance.
- In some example embodiments, one of more variations may be made as well as described in the detailed description below and/or as described in the following features. The critical transconductance may be determined based on the equation: gm,crit=Ci/CeffZo wherein Ci is an input capacitance, Ceff is an effective capacitance, and Zo is a characteristic impedance. The one or more active devices may include a bipolar junction transistor, a field effect transistor, a metal oxide semiconductor field effect transistor, a metal semiconductor field effect transistor, and a high electron mobility transistor.
- In some example embodiments, a method may be provided. The method may include determining a bias point for an active device in a radio frequency power amplifier, wherein the bias point is selected to cause a transconductance of the active device to be above a critical transconductance value; determining a parasitic capacitance between an input of the active device and an output of the active device; selecting one or more reactive components in a feedback circuit of the radio frequency power amplifier to produce a parallel tank circuit, the one or more reactive components including the parasitic capacitance; and determining one or more values for the one or more reactive components in the parallel tank circuit to cause an input impedance match and an output impedance match.
- In some example embodiments, one of more variations may be made as well as described in the detailed description below and/or as described in the following features. The feedback network may include an inductor connected between an input port and an output port, an input capacitor connected between the input port and the base of the transistor, a parasitic capacitance of the transistor between the base of the transistor and the collector of the transistor, and an output capacitor connected between the output port and the collector of the transistor. The feedback network may include a feedback capacitor in parallel with the parasitic base-collector capacitance of the transistor, and/or a parasitic capacitance of the transistor between the base of the transistor and the emitter of the transistor.
- It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive. Further features and/or variations may be provided in addition to those set forth herein. For example, the implementations described herein may be directed to various combinations and subcombinations of the disclosed features and/or combinations and subcombinations of several further features disclosed below in the detailed description.
- The accompanying drawings, which are incorporated in and constitute a part of this specification, show certain aspects of the subject matter disclosed herein and, together with the description, help explain some of the principles associated with the disclosed implementations. In the drawings,
-
FIG. 1 is a diagram showing a source, a load, and an example of a power amplifier, in accordance with some implementations; -
FIG. 2A depicts a diagram showing an example of a single stage radio frequency (RF) power amplifier incorporating an active device, in accordance with some implementations; -
FIG. 2B depicts a diagram showing an example of a single stage RF power amplifier incorporating a bipolar junction transistor as the active device, in accordance with some implementations; -
FIG. 3A shows an alternating circuit (AC) circuit diagram of a single stage RF power amplifier, in accordance with some implementations; -
FIG. 3B shows a small-signal AC circuit model including a feedback network, in accordance with some implementations; -
FIG. 4 shows a process for producing an RF power amplifier stage, in accordance with some implementations; -
FIG. 5A shows the two-port S-parameters of a circuit similar toFIG. 2B ; -
FIG. 5B shows the effect of the voltage divider created by capacitances, such as input capacitance Ci and base-emitter capacitance Cπ, on the collector current as a function of input power, in accordance with some implementations; -
FIG. 6 is a circuit diagram showing a three-stage power amplifier, in accordance with some implementations; and -
FIG. 7 is a photograph of a chip containing a three-stage power amplifier, in accordance with some implementations. - When practical, similar reference numbers denote similar structures, features, or elements.
-
FIG. 1 depicts an example of a radiofrequency power amplifier 150 including a radio frequency (RF)source 110 having asource impedance Z s 120, afeedback network 190, anactive device 195, and aload 180 having aload impedance Z L 180, in accordance with some example implementations. TheRF source 110 may have aninput power P i 130 available to supply toRF power amplifier 150 atinput port 135. In the example ofFIG. 1 , theRF source 110 may provide a radio frequency carrier that is modulated to carry information. In some example implementations, theRF source 110 may generate a RF carrier having a frequency between 5 GHz and 150 GHz including the millimeter-wave bands, such as Q-band (30-50 GHz), V-band (50-75 GHz), and W-band (75-110 GHz), although other frequency ranges may be used as well. Theradio frequency amplifier 150 may also include an input impedance Zin 140 (e.g., in ohms), anoutput power P o 170 delivered atoutput port 165 to loadimpedance Z L 180, and an output impedance Zout 160 (e.g., in ohms).RF power amplifier 150 amplifiesinput power P i 130 and delivers output power Po to loadimpedance Z L 180. - In some example implementations, RF
power amplifier circuit 100 may be configured, such that a source impedance, an input impedance, an output impedance, and a load impedance are each determined so as to substantially match an impedance value (referred to herein as a characteristic impedance). For example, source impedance Zs 120 and load impedance ZL 180 may each be selected to have a value equal to about a characteristic impedance, Zo. Moreover,RF power amplifier 150 may be configured so thatactive device 195 andfeedback network 190 are selected to provideinput impedance Z in 140 equal to about the characteristic impedance andoutput impedance Z out 160 equal to about the characteristic impedance. In some example implementations, the characteristic impedance may be equal to about 50 ohms, although other impedance values may be used as well. - In some example implementations, the
feedback network 190 combined with theactive device 195 may be configured to nullify the deleterious effects of the parasitic elements ofactive device 195 in order to provide higher output power, greater amplifier efficiency, and/or provide good input and output matching. The combination of theinput impedance Z in 140 and thesource impedance Z s 120 may determine a reflection coefficient (related to as the Scattering parameter S11, when theinput port 135 is port 1). When the reflection coefficient is about equal to zero or nearly zero (e.g., less than or equal to about 0.1 or equivalently −10 dB), the impedances, such assource impedance Z s 120 and RF power amplifier input impedance Zin, may be considered well matched. - In some example implementations, the parasitic elements of one or more active devices, such as
active device 195, may be combined with external components, such as for example inductors and capacitors selected to form a tank circuit resonant at a predetermined frequency. The tank circuit may include parasitic components of theactive device 195, such as a parasitic base-collector capacitance, Cμ, if the active device is a bipolar junction transistor. When anactive device 195 is biased so that the transconductance of theactive device 195 is above a certain value, such as for example a critical value, then at, or near, the resonant frequency of the tank circuit, theRF power amplifier 150 may, in some example implementations, achieve a higher output power before going into gain compression, when compared to an active device without the parallel tank circuit. Transconductance values above the critical value may also provide higher output power before going into compression. However, selecting a transconductance value below the critical value may result in poorer input and output matching and degraded linearity and output compression at a lower output power compared to the matching and linearity when the transconductance is above the critical value. Along with the increased output power, thefeedback network 190 and theactive device 195 may also cause an impedance match at theinput port 135 between theinput impedance Z in 140 andsource impedance Z s 120. And, thefeedback network 190 and theactive device 195 having a transconductance at, or above, the threshold value may cause an impedance match at theoutput port 165 between theoutput impedance Z out 160 andload impedance Z L 180. - In some example implementations, the
RF amplifier 150 may include one or more active devices and one or more passive devices. Examples of active devices include a bipolar junction transistor (BJT), a heterojunction bipolar transistor (HBT), a field effect transistor (FET), a metal oxide semiconductor field effect transistor (MOSFET), a metal semiconductor field effect transistor (MESFET), and a high electron mobility transistor (HEMT), although other active devices may be included inRF amplifier 150 as well. In some example implementations, theactive device 195 is chosen to be a bipolar junction transistor produced in a 120 nm silicon-germanium (SiGe) BiCMOS process, although other processes may be used as well. Examples of passive devices include an inductor, a capacitor, and a transmission line, although other types of passive devices may be included inRF power amplifier 150 as well. Moreover, the devices used in thepower amplifier 150 may be implemented in a semiconductor processes, such as for example silicon (Si), silicon germanium (SiGe), gallium arsenide (GaAs), indium phosphide (INP), gallium nitride (GaN) as well as others. In some example implementations, the semiconductor process may allow for multiple active device types, such as in a bipolar/complementary metal oxide semiconductor (BiCMOS) process that allows for biploar transistors and complementary metal oxide semiconductor (CMOS) devices to be produced using the same process. -
RF power amplifier 150 may include a single amplification stage, such asamplifier stage 152 or multiple amplifier stages. Multiple amplifier stages may be cascaded with theinput port 135 connected to the input to the first stage and the output of the first stage being connected to the input of the next stage, and so on, up to the last stage where the output of the last stage is connected tooutput port 165. Multiple amplifier stages may also be interconnected in parallel where the outputs of multiple stages are combined together atoutput port 165. A combination of cascaded amplifier stages and parallel amplifier stages may also be used. -
FIG. 2A depicts an example of a single RFpower amplifier stage 200A. The RF power amplifiersingle stage 200A may include aninput port 260 for receiving a signal for amplification, afeedback network 210, anactive device 220A, and anoutput port 270 for supplying an amplified output signal. In some example implementations, theactive device 220A may be an active device configured in accordance with a semiconductor process as noted above (e.g., a bipolar junction transistor implemented in a BiCMOS process, and the like). Thefeedback network 210 may include aninput capacitor C i 230, aninductor L 240, anoutput capacitor C o 250, andfeedback capacitor C μ est 255. - In
stage 200A,input capacitor C i 230 may connectinput port 260 to the gate/base/input ofactive device 220A,inductor L 240 may connectinput port 260 tooutput port 270,output capacitor C o 250 may connectoutput port 270 to the drain/collector/output ofactive device 220A, andfeedback capacitor C μ ext 255 may connect theinput port 260 to theoutput port 270. The source/emitter ofactive device 200A may be connected to a ground (e.g., an alternating current (AC) ground). -
FIG. 2B depicts an example of a single RFpower amplifier stage 200B implemented using abipolar junction transistor 220B. For example, instage 200B,input capacitor C i 230 may connectinput port 260 to the base ofbipolar junction transistor 220B,inductor L 240 may connectinput port 260 tooutput port 270,output capacitor C o 250 may connectoutput port 270 to the collector ofbipolar junction transistor 220B, andfeedback capacitor C μ ext 255 may connect theinput port 260 to theoutput port 270. The parasitic capacitance ofbipolar junction transistor 220B, may also connect theinput port 260 to theoutput port 270. The emitter ofbipolar junction transistor 220B may be connected to AC ground. A parallel tank circuit may be formed frominductor L 240 in parallel with the series combination ofinput capacitance C i 230,output capacitance C o 250, and the parallel combination of parasitic base-collector capacitance Cμ ofbipolar junction transistor 220B in parallel with feedback capacitance Cμ ext ext 255. In some example implementations,feedback capacitance C μ ext 255 may have a zero or nearly zero capacitance value. - The parasitic base-collector capacitance Cμ of
bipolar junction transistor 220B may be an equivalent value determined based on the configuration ofbipolar junction transistor 220B and the process selected to fabricatebipolar junction transistor 220B. The value of parasitic base-collector capacitance Cμ may be determined based on the device physics, or measured. -
FIG. 3A depicts an example of a single stageRF power amplifier 300, which is similar in some respects to thepower amplifier 200B inFIG. 2B . Theamplifier 300 may include aninput source 110 with asource impedance Z s 120 about equal to a characteristic impedance Zo and aload impedance 180 about equal to a characteristic impedance Zo. For simplicity, FIG. 3A depicts an AC equivalent circuit without a bias network, although it may be presumed that thebipolar transistor 220B is biased to a transconductance above a critical transconductance. -
FIG. 3B depicts an example of a small-signal circuit model 305 representation of the single stageRF power amplifier 300 atFIG. 3A . The small-signal circuit model 305 may include the active device model parameters, such as parasitic base-collector capacitance C μ 315,transconductance g m 330,substrate capacitance C s 340, and base-emitter capacitance C π 320. The parallelresonant tank circuit 310 may include a feedback network, such asfeedback network 210 and the parasitic base-collector capacitance (or gate-drain capacitance) of the active device, such as base-collector capacitance C μ 315. - When the active device transconductance, gm, is selected to be greater than the critical transconductance (e.g., gm≧gm,crit), the following device behavior may be observed:
-
ωo=1/√{square root over ((LC eff))} Equation 1A, - wherein
-
- ωo is the resonant frequency of the parallel tank circuit,
- L represents the
inductance L 240 of the feedback network, and - Ceff represents the effective capacitance of the series combination of
- input capacitance Ci, output capacitance Co, and the total base-collector capacitance (Cμ.+Cμ ext).
- And, Ceff may be determined in accordance with the following:
-
- In some representations, the sum of Cμ. and Cμ ext is referred to as just Cμ.
- And, the voltage gain, Av, represents the voltage gain of the power amplifier, wherein Av may be determined in accordance with the following:
-
A v =C i /C μ Equation 1C. - At the resonant frequency, ωo, of the
parallel tank circuit 310, thetransconductance 330 of theactive device 220B and theparallel tank circuit 310 may cause theinput impedance Z in 140 to be about equal to the characteristic impedance Zo and may cause theoutput impedance Z out 160 to be about equal to characteristic impedance Zo. - With the
input port 260 asport 1 and theoutput port 270 asport 2, the two-port S-parameters of the small-signal circuit model shown inFIG. 3B are as follows: -
- wherein α=(LCi 2Zo)/(2), β=(LCi 2)/(2Ceffg
m Zo)+LCi, and γ=(Ci 2Zo)/(2Ceff)+(L)/(2Zo) may determine the frequency dependency of the S-parameters. - The input and output return loss, S11 and S22, are expressed in Equation 3A. Instead of using separate input matching and output matching circuits, the tank circuit including
feedback network 210 andactive device 220A/B may be configured to provide both input and output matching. - S11 and S22 are minimized at a frequency, ωnotch expressed as follows:
-
ωnotch=√{square root over ((1)/(C eff L)−(1)/(C i Z o)2)}{square root over ((1)/(C eff L)−(1)/(C i Z o)2)}{square root over ((1)/(C eff L)−(1)/(C i Z o)2)}{square root over ((1)/(C eff L)−(1)/(C i Z o)2)}Equation 3. - which is slightly below ωo. When the voltage gain is high,
-
- ωnotch approaches ωo.
- For values of transconductance gm below the critical transconductance gm,crit, at resonant frequency ωo, the input impedance is high compared to characteristic impedance Zo, since β dominates both the numerator and denominator in Equation 2A. In this case, the
parallel tank circuit 310 andtransconductance 330 simplify to a parallel LC circuit which is an open circuit at resonant frequency ωo. For larger values of the transconductance gm, the real part of the input (and output) impedance decreases and the input and output impedances converge to the following: -
- While the real part of the impedance approaches the characteristic impedance value Zo, the imaginary part is inversely proportional to the input capacitance Ci. Thus, the input and output matching may, in some example implementations, improve as the
input capacitance C i 230 becomes larger. The required transconductance gm for a return loss better than 10 dB may be defined as the critical transconductance as follows: -
- Increasing both the transconductance gm and the ratio of input capacitance to the base-collector capacitance (e.g., Ci/Cμ) may, in some example implementations, improve the return loss. For an input-output capacitance ratio equal to about 2 (e.g., Ci/Cμ=2), increasing the transconductance gm above the critical transconductance value gm,crit, may not substantially improve (e.g., increase) the return loss.
- From Equation 2B, the peak value of the gain, S21, occurs at a frequency, such as ωpeak=√{square root over ((1)/CeffL)+(1)/((CiZo)2))}{square root over ((1)/CeffL)+(1)/((CiZo)2))}{square root over ((1)/CeffL)+(1)/((CiZo)2))}, when the denominator of Equation 2B is minimized. Because of the numerator in Equations 2B and 2C, the reverse voltage gain, S12, is much smaller than the forward voltage gain S21. As such, the isolation between the
input port 135 and theoutput port 165 increases with the gain in some example implementations. For a transconductance much greater than the critical transconductance, Beta approaches the value of the product of theinductance L 240 and input capacitance Ci 230 (e.g., gm>>gm,crit, β→LCi), simplifying the following equation (2β−LCi)s2+(L)/(Zo)s+1to LCis2+(L)/(Zo)s+1, which yields a local minimum at a frequency ω=√{square root over ((1)/(LCi))}{square root over ((1)/(LCi))}. Consequently, the pole at this frequency, ω=√{square root over ((1)/(LCi))}{square root over ((1)/(LCi))}, from 1+LCis2 is cancelled due to the following: (2β−LCi)s2+(L)/(Zo)s+1. - The frequency where S11 and S22 are at a minimum (which is also referred to as a notch frequency, ωnotch) and the frequency where S21 peaks, ωpeak, are related to the resonant frequency ωo as follows:
-
ωnotch 2+ωpeak 2=ωo 2 Equation 6. - For high input capacitance Ci values, the notch frequency and peak frequency approach the resonant frequency ωo. At the resonant frequency, ωo, Equations 2B and 2C may be simplified as follows:
-
- The quality factor, Q, may be expressed as follows:
-
Q=(Z o)/(ωo L)=(Z o)/(√{square root over ((L)/(C eff))}{square root over ((L)/(C eff))}) Equation 8. - Since the quality factor Q of the of the parallel tank circuit in Equation 8 may be inversely proportional to the value of the
inductor L 240, a smaller inductor value may make the parallel tank more narrowband, while a larger value may make the parallel tank more broadband. A smaller value of the effective capacitance Ceff may make the parallel tank more wideband, while a larger capacitor value may make the parallel tank more narrowband. The return losses, S11 and S22, may be broadened using balanced architectures or staggering the return loss across multiple power amplifier stages. - The above disclosure is based on the base-emitter capacitance Cπ and the substrate capacitance Cs being zero or about zero. Generally, the substrate capacitance Cs may be relatively small and thus has minor effects. The base-emitter capacitance Cπ may be an order of magnitude larger than the input capacitance Ci due to the value of the transconductance gm. The base-emitter capacitance Cπ reduces the gain, S21. However, this reduction in gain may have the beneficial effect of keeping the active device out of compression at higher output voltage swings.
-
FIG. 4 depicts aprocess 400 for producing an RF power amplifier stage, in accordance with some example implementations. - At 410, a processor may determine, for the RF amplifier, one or more of the following amplifier parameters: a device technology (e.g., silicon-germanium, gallium-arsenide, etc), an active device structure (e.g., bipolar junction transistor, field-effect transistor, etc.), a quantity of power amplifier stages, an operating frequency, a gain, an output power, and any other parameters which may be determined or optimized by a processor. Some of these parameters may be dictated based on, for example, the intended application, such as a satellite communications application for a particular satellite. For example, an RF power amplifier for millimeter-wave satellite communications may require 200 milliwatts (23 dBm) of output power at 38 Gigahertz. In this example, the parameters which may be determined at 410 may thus include the device technology, active device structure, the quantity of stages, and the gain.
- At 420, a bias point for the active device, such as
active deice 195, may be selected, such that the active device's transconductance (which is determined by the bias point) is about greater than a critical transconductance value, as described above with respect toFIGS. 3A and 3B andEquation 5. For example, a processor may select a bias point representative of a transconductance value greater than the critical value given inEquation 5. The processor may select the bias point by selecting the bias voltage that corresponds to the selected transconductance. However, selecting a transconductance value below the critical value may result in poorer input and output matching and/or degraded linearity compared to the matching and linearity realized when the transconductance is above the critical value. The bias point may be implemented by setting a direct current (DC) base/source/input voltage. - At 430, the values of the components of a circuit model of the active device may be determined. For example, the values of the base-
emitter capacitance C π 320, and the base-collector capacitance C μ 315 may depend on the bias point ofbipolar junction transistor 220B. The values of the base-emitter capacitance C π 320 andC μ 315 may be determined after the bias point is determined. - At 440, the values of the components of the feedback network, such as
feedback network 210 with components includinginput capacitance C i 230,output capacitance C o 250,inductance L 240, andfeedback capacitance C μ ext 255 may be chosen. The values of these components may be chosen to provide, when theactive device transconductance 330 is substantially above the critical transconductance in accordance withEquation 5, a good input match atinput port 260, a good output match atoutput port 270, and aparallel tank 310 with a predetermined resonant frequency, ωo. For example, a processor may selectinput capacitance C i 230,output capacitance C o 250,inductance L 240,feedback capacitance C μ ext 255, and a bias point to provide a transconductance above the critical transconductance in accordance withEquation 5, and good input and output matching in accordance with Equation 2A. In the example ofFIG. 2B , the feedback network includes the parallel tank includinginductor L 240 in parallel with the series combination ofinput capacitance C i 230, base-collector capacitance C μ 315 from the bipolar junction transistor small-signal circuit model,output capacitance C o 250, andfeedback capacitance C μ ext 255. In some implementations the predetermined resonant frequency of the parallel tank circuit is the center of the operating frequency of the RF power amplifier. - In some example implementations,
process 400 may be repeated for each stage of a multistage RF power amplifier and/or repeated for an individual stage. In any case,process 400 may be repeated in order to optimize the determination of the active bias point and the feedback network. - Referring again to
FIG. 3A , the large-signal behavior of RFpower amplifier stage 300 inFIG. 3A may depend on the biasing ofbipolar junction transistor 220B. With a selection of the transconductance gm to be above the critical transconductance in gm,crit,Equation 5, the selection ofinductor L 240,input capacitance C i 230,output capacitance C o 250, andfeedback capacitance C μ ext 255, the circuit inFIG. 3B may exhibit one or more of the following attributes at the resonant frequency of the parallel tank circuit 310: - 1) The power delivered from the source to the input of
amplifier stage 300, for exampleinput power P i 130, may be maximized when theinput impedance Z i 140 is equal to about thesource impedance Z s 120, wherein theinput power P i 130 may be expressed in terms of the input current ii and the characteristic impedance Zo, or in terms of the input source voltage and characteristic impedance Zo according to the following equation Pin=(ii 2(ωo)·Zo)/(2)=(Vs 2)/(4Zo); - 2) The power delivered to the load is
output power P o 170, wherein theoutput power 170 may be expressed in terms of the output current io and the characteristic impedance Zo, or in terms of the input current ii, the voltage gain Av and the characteristic impedance Zo according to the following equation: Po=(io 2(ωo)·Zo)/(2)=(Av 2·ii 2(ωo)·Zo/(2), wherein thesource impedance Z s 120 is equal to theload impedance Z L 180, which are both equal to the characteristic impedance Zo; - 3) The transducer power gain GT may be represented as a ratio of the
output power P out 170 delivered to the load to the input power available at thesource P in 130 or as the voltage gain squared according to the following equation: GT−(Pout)/(Pin)=Av 2, which depends on theinput capacitance C i 230, and base-collector capacitance,C μ 315; - 4) The collector efficiency η may be represented as a ratio of the
output power P out 170 to the direct current power supplied to the power amplifier Pdc or in terms of the peak output current io, characteristic impedance Zo, direct current collector bias voltage Vcc, and collector current Ic according to the following equation: -
η=(P out)/(P dc)=(i o 2 Z o)/(2V cc I c) Equation 9; and - 5) The maximum collector efficiency may be expressed in terms of the voltage gain Av as follows:
-
- While the efficiency, η, in Equation 9 may be influenced by the RF power amplifier class (a class A amplifier, a class B amplifier, a class AB amplifier, and the like), the
RF power amplifier 150 may achieve, in some example implementations, higher efficiency at high input power levels because in part the load line matching does not require a high-Q impedance transformation. Instead of using separate input matching and output matching circuits,tank circuit 310 may provide both input and output matching. - Additionally, the base-emitter capacitance,
C π 320, may substantially linearize the power amplifier input-output transfer function, which may provide for a higher input power before the output goes into compression. This linearization may be due to a voltage divider formed byinput capacitor C i 230 and base-emitter capacitance C π 320, wherein input voltage Vi is the input voltage atinput port 260 and base-emitter voltage Vπ is the voltage between the base and emitter ofbipolar junction transistor 220B. The voltage divider causes a portion of the input voltage Vi to be across the input capacitor Ci and the remaining portion of the input voltage Vi to be across the base-emitter (equal to Vπ) ofbipolar junction transistor 220B. The base-emitter voltage Vπ may be represented in terms of a ratio involving theinput capacitance C i 230 and the base-emitter capacitance C π 320 in Equation 11. With the base-collector capacitance C μ 315 andfeedback capacitance C μ ext 255 inFIG. 3B being small compared to Ci and Cπ, they may be neglected in the representation of the voltage divider. -
FIG. 5A shows a two-port S-parameters of a circuit similar toFIG. 2B . Shown are the input match S11, the output match S22, the gain S21 and the reverse gain S12. The minimum of S11 and S22 at the notch frequency,ω notch 502, related toEquations 3 and 6 is also shown. The maximum of S21 at the peak frequency ,ω peak 501, related to Equation 6 is also shown. -
FIG. 5B shows the effect of the voltage divider. The voltage divider causes the RF power amplifier disclosed here to delay compression of the output signal atoutput port 270 to a higher power input signal atinput port 260 compared to traditional design techniques. - The plot at 510 depicts the bipolar junction transistor collector current as a function of input power, in accordance with some example implementations. The
voltage V π 510A at the base of the active device, such asbipolar junction transistor 220B, may be the result of the input voltage Vi divided across input capacitance Ci and base-emitter capacitance Cπ wherein, -
- In some example implementations, the effect of the voltage divider may be to linearize the transfer function of the output current as a function of input power. This may extend the range of input power before the output of the power amplifier goes into compression. For example, the effect of the voltage divider may be to increase the input power corresponding to the 1 dB compression point of the
output 270. The voltage divider may, in some example implementations, improve the 1 dB compression point by a factor involving the ratio of the base-emitter capacitance to the input capacitance expressed as follows: -
(1+(Cπ)/(Ci))Equation 12. - The plot at 520 depicts the input power versus the collector current without input capacitor Ci and thus no voltage divider. Without input capacitor Ci, the voltage across the base-
emitter V π 520A of the transistor is equal to the input voltage Vi. The result of not having the voltage divider is to produce more non-linearity in the input power versus collector current transfer function thus reducing the input power corresponding to the 1 dB compression point of the output. -
FIG. 6 is a circuit diagram showing a three-stage power amplifier 600. -
Stage 1 at 610 receives an input signal frominput port 135 throughinductor L1 611. Thestage 1 parallel tank circuit includesinductor L2 612,capacitor C1 613,capacitor C2 615 and the base-collector capacitance (Cμ) ofbipolar junction transistor 616.Bipolar junction transistor 616 is biased at 618 by a bias network such asbias network 601. In the implementation shown instage 1 610, inductor L1=150 pH (picohenries), inductor L2=320 pH, capacitors C1=C2=220 fF (femtofarads), andbipolar junction transistor 616 has a size of 36 microns. -
Stage 2 at 620 receives an input signal fromstage 1 throughinductor L3 621. Thestage 2 parallel tank circuit includesinductor L4 622,capacitor C3 623,capacitor C4 625 and the base-collector capacitance (Cμ) ofbipolar junction transistor 626.Bipolar junction transistor 626 is biased at 628 by a bias network such asbias network 601. In the implementation shown instage 2 620, inductor L3=150 pH, inductor L4=240 pH, capacitors C3=C4=200 fF, andbipolar junction transistor 626 has a size of 72 microns. -
Stage 3 at 620 receives an input signal fromstage 2 throughinductor L5 631. Thestage 3 parallel tank circuit includesinductor L6 632,capacitor C5 633,capacitor C6 635 and the base-collector capacitance (Cμ) ofbipolar junction transistor 636.Bipolar junction transistor 636 is biased at 638 by a bias network such asbias network 601. In the implementation shown instage 3 630, inductor L5=150 pH, inductor L6=180 pH, capacitors C5=C6=180 fF, andbipolar junction transistor 626 has a size of 144 microns. -
Output port 165 provides the output ofRF power amplifier 600 fromstage 3 throughinductor L7 641 which has a value of 150 pH. The component values inpower amplifier 600 are representative of a particular implementation where the values a chosen in accordance with the foregoing disclosure. Any number of other implementations are also possible that satisfy the foregoing considerations. -
FIG. 7 shows a drawing 700 of a die containing a circuit similar to the circuit ofFIG. 6 implemented using a 120 nm SiGe BiCMOS process. The circuit measures 1160 microns×900 microns. The implementation at 700 has three RF power amplifier stages 610, 620, and 630, which are similar in some respects to the three stages atFIG. 6 . - The input port to the RF power amplifier in 700 is a coplanar waveguide ground-signal-ground structure shown in 135A and 135B. The physical inductors of
FIG. 6 are depicted asinductors L1 611,L2 612,L3 621,L4 622,L5 631,L6 632 andL7 641. The power amplifier output port is depicted at 165A and 165B.Capacitors C1 613,C2 615,C3 623,C4 625,C5 633, andC6 635 are also present but not labeled as arebipolar junction transistors - The nominal collector current bias for all three stages is 4 mA at 2.4-V collector voltage. This RF power amplifier built in accordance with the foregoing may, in some example implementations, achieve a maximum power added efficiency of 20% with a saturated output power of 21.1 dBm; a saturated output power of 23 dBm with a collector voltage of 3 V; and/or a 3-dB bandwidth of 5-GHz from 36 to 41 GHz, although other implementations may realize other performance values as well.
Claims (14)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US14/356,677 US9160286B2 (en) | 2011-11-09 | 2012-11-09 | Radio frequency power amplifiers |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US201161557866P | 2011-11-09 | 2011-11-09 | |
PCT/US2012/064484 WO2013071133A2 (en) | 2011-11-09 | 2012-11-09 | Radio frequency power amplifiers |
US14/356,677 US9160286B2 (en) | 2011-11-09 | 2012-11-09 | Radio frequency power amplifiers |
Publications (2)
Publication Number | Publication Date |
---|---|
US20140320212A1 true US20140320212A1 (en) | 2014-10-30 |
US9160286B2 US9160286B2 (en) | 2015-10-13 |
Family
ID=48290771
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US14/356,677 Active US9160286B2 (en) | 2011-11-09 | 2012-11-09 | Radio frequency power amplifiers |
Country Status (2)
Country | Link |
---|---|
US (1) | US9160286B2 (en) |
WO (1) | WO2013071133A2 (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170040958A1 (en) * | 2014-07-11 | 2017-02-09 | Skyworks Solutions, Inc. | Amplifier with termination circuit and resonant circuit |
WO2018045093A1 (en) * | 2016-08-30 | 2018-03-08 | Macom Technology Solutions Holdings, Inc. | Driver with distributed architecture |
US10630052B2 (en) | 2017-10-04 | 2020-04-21 | Macom Technology Solutions Holdings, Inc. | Efficiency improved driver for laser diode in optical communication |
CN111416586A (en) * | 2020-04-03 | 2020-07-14 | 杭州易百德微电子有限公司 | Load structure and radio frequency amplifier formed by same |
CN112910420A (en) * | 2021-01-18 | 2021-06-04 | 温州大学 | High-linearity radio frequency power amplifier |
US11095318B2 (en) * | 2019-07-24 | 2021-08-17 | Samsung Electronics Co., Ltd. | Transceiver using active device array and antenna module including the same |
US11463177B2 (en) | 2018-11-20 | 2022-10-04 | Macom Technology Solutions Holdings, Inc. | Optic signal receiver with dynamic control |
US11658630B2 (en) | 2020-12-04 | 2023-05-23 | Macom Technology Solutions Holdings, Inc. | Single servo loop controlling an automatic gain control and current sourcing mechanism |
US12013423B2 (en) | 2020-09-30 | 2024-06-18 | Macom Technology Solutions Holdings, Inc. | TIA bandwidth testing system and method |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11815588B2 (en) * | 2019-07-05 | 2023-11-14 | University Of Electronic Science And Technology Of China | Room-temperature semiconductor maser and applications thereof |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8456237B2 (en) * | 2011-03-23 | 2013-06-04 | Integrated Device Technology, Inc. | Low noise variable gain amplifier utilizing variable feedback techniques with constant input/output impedance |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090231042A1 (en) | 2007-09-05 | 2009-09-17 | Pengcheng Jia | Broadband Power Amplifier with A High Power Feedback Structure |
CN102754334B (en) | 2010-01-04 | 2016-03-02 | 马维尔国际贸易有限公司 | With the power amplifier of feedback impedance for stable output |
-
2012
- 2012-11-09 WO PCT/US2012/064484 patent/WO2013071133A2/en active Application Filing
- 2012-11-09 US US14/356,677 patent/US9160286B2/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8456237B2 (en) * | 2011-03-23 | 2013-06-04 | Integrated Device Technology, Inc. | Low noise variable gain amplifier utilizing variable feedback techniques with constant input/output impedance |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10135408B2 (en) * | 2014-07-11 | 2018-11-20 | Skyworks Solutions, Inc. | Amplifier with termination circuit and resonant circuit |
US20170040958A1 (en) * | 2014-07-11 | 2017-02-09 | Skyworks Solutions, Inc. | Amplifier with termination circuit and resonant circuit |
TWI750216B (en) * | 2016-08-30 | 2021-12-21 | 美商Macom技術方案控股公司 | Driver with distributed architecture |
WO2018045093A1 (en) * | 2016-08-30 | 2018-03-08 | Macom Technology Solutions Holdings, Inc. | Driver with distributed architecture |
US10263573B2 (en) | 2016-08-30 | 2019-04-16 | Macom Technology Solutions Holdings, Inc. | Driver with distributed architecture |
US10630052B2 (en) | 2017-10-04 | 2020-04-21 | Macom Technology Solutions Holdings, Inc. | Efficiency improved driver for laser diode in optical communication |
US11463177B2 (en) | 2018-11-20 | 2022-10-04 | Macom Technology Solutions Holdings, Inc. | Optic signal receiver with dynamic control |
US11095318B2 (en) * | 2019-07-24 | 2021-08-17 | Samsung Electronics Co., Ltd. | Transceiver using active device array and antenna module including the same |
US11955997B2 (en) | 2019-07-24 | 2024-04-09 | Samsung Electronics Co., Ltd. | Transceiver using active device array and antenna module including the same |
CN111416586A (en) * | 2020-04-03 | 2020-07-14 | 杭州易百德微电子有限公司 | Load structure and radio frequency amplifier formed by same |
US12013423B2 (en) | 2020-09-30 | 2024-06-18 | Macom Technology Solutions Holdings, Inc. | TIA bandwidth testing system and method |
US11658630B2 (en) | 2020-12-04 | 2023-05-23 | Macom Technology Solutions Holdings, Inc. | Single servo loop controlling an automatic gain control and current sourcing mechanism |
CN112910420A (en) * | 2021-01-18 | 2021-06-04 | 温州大学 | High-linearity radio frequency power amplifier |
Also Published As
Publication number | Publication date |
---|---|
US9160286B2 (en) | 2015-10-13 |
WO2013071133A2 (en) | 2013-05-16 |
WO2013071133A3 (en) | 2013-07-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9160286B2 (en) | Radio frequency power amplifiers | |
Chappidi et al. | Globally optimal matching networks with lossy passives and efficiency bounds | |
US7489201B2 (en) | Millimeter-wave cascode amplifier gain boosting technique | |
Sato et al. | 68–110-GHz-band low-noise amplifier using current reuse topology | |
Ng et al. | A 20-watt Ka-band GaN high power amplifier MMIC | |
US20120242410A1 (en) | Bipolar stacked transistor architecture | |
US7466205B1 (en) | Ultra-wideband low noise amplifier and amplification method thereof | |
Ma et al. | A 30GHz 2dB NF low noise amplifier for Ka-band applications | |
Wang et al. | A high gain broadband LNA in GaAs 0.15-μm pHEMT process using inductive feedback gain compensation for radio astronomy applications | |
CN107659278A (en) | A kind of Ka wave bands SiGe BiCMOS radio-frequency power amplifiers | |
Nguyen et al. | A 7–115-GHz distributed amplifier with 24-dBm output power using quadruple-stacked HBT in InP | |
Friesicke et al. | A linear 4W power amplifier at K-band using 250nm AlGaN/GaN HEMTs | |
Gashi et al. | Broadband and high-gain 400-GHz InGaAs mHEMT medium-power amplifier S-MMIC | |
John et al. | 670-GHz cascode circuits based on InGaAs metamorphic high-electron-mobility transistors | |
CN114710127A (en) | W-waveband high-gain power amplifier circuit | |
CN114094955A (en) | W-waveband high-gain low-noise amplifier circuit | |
Hamada et al. | 230–305 GHz,> 10-dBm-output-power wideband power amplifier using low-Q neutralization technique in 60-nm InP-HEMT technology | |
Lee et al. | A W-band amplifier with a new wide-band interstage matching technique using self-resonance of a microstrip-coupled line | |
Jensen et al. | Multifinger InP HBT's in transferred-substrate technology for 100 GHz power amplifiers | |
Jarndal et al. | On designing of a broadband GaN low-noise amplifier for WiMax applications | |
Do et al. | 60 GHz SiGe LNA | |
Jiang et al. | Cryogenic L-band GaAs HEMT LNA for the Square Kilometer Array | |
Hu et al. | Analysis and design of a stacked power amplifier with 196% fractional bandwidth using equivalent circuit model | |
Koh et al. | Broadband linearization of InGaP/GaAs HBT power amplifier | |
Mehrotra et al. | Design and Analysis of a Broadband Low Noise Amplifier using 0.25 μm GaAs pHEMT Process |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: THE REGENTS OF THE UNIVERSITY OF CALIFORNIA, CALIF Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:KALANTARI, NADER;BUCKWALTER, JAMES;SIGNING DATES FROM 20111118 TO 20111128;REEL/FRAME:035876/0979 |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
MAFP | Maintenance fee payment |
Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YR, SMALL ENTITY (ORIGINAL EVENT CODE: M2551); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY Year of fee payment: 4 |
|
MAFP | Maintenance fee payment |
Free format text: PAYMENT OF MAINTENANCE FEE, 8TH YR, SMALL ENTITY (ORIGINAL EVENT CODE: M2552); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY Year of fee payment: 8 |